Aggression and Violent Behavior 50 (2020) 101343

Contents lists available at ScienceDirect

Aggression and Violent Behavior

journal homepage: www.elsevier.com/locate/aggviobeh

Cyberbullying: Concepts, theories, and correlates informing evidence-based T

best practices for prevention
⁎
Nadia S. Ansary
Rider University, United States of America

A R T I C LE I N FO A B S T R A C T

Keywords: Emerging evidence has revealed that many characteristics of cyberbullying—its definition, prevalence rates, risk
Cyberbullying and protective factors, outcomes, and prevention strategies—are related and yet somewhat unique from tradi-
Cybervictimization tional bullying. The ubiquity of technology in the lives of youth presents an opportunity for individuals to
Electronic bullying intentionally and repetitively harm others, 24 h per day, sometimes with complete anonymity, and often without
Prevention
consequence. This is concerning given the high rates of psychopathology associated with cybervictimization,
Intervention
over and above, traditional bullying. Given the current state of the field, this literature review provides a critical
synthesis of the extant knowledge concerning (1) a definition of cyberbullying; (2) theories explaining cyber-
bullying; (3) prevalence rates; (4) a brief developmentally-focused overview of adolescents and their online use;
(5) risk and protective factors; (6) negative psycho-social outcomes, over and above traditional bullying; and (7)
a brief overview of prevention and intervention programming with information for key stakeholders.
Implications and future directions are discussed.

While some researchers view cyberbullying as not profoundly dis-
tinct from traditional bullying (Olweus & Limber, 2018; see
Vaillancourt, Faris, & Mishna, 2017), the nuanced differences between
the two suggest that generalizing findings from traditional bullying to
cyberbullying may be inadequate (Savage & Tokunaga, 2017). Growing
evidence reveals that many facets of cyberbullying—from its definition,
prevalence rates, risk and protective factors, outcomes, and prevention
strategies—are related and yet somewhat distinct from traditional
bullying (Espelage & Hong, 2017; Kowalski, Limber, & McCord, 2019;
Patchin & Hinduja, 2012; Smith et al., 2008; Vaillancourt et al., 2017;
Zych, Farrington, & Ttofi, 2019). Cyberspace makes bullying easier,
more accessible, and inflicts greater harm than traditional bullying (see
Vaillancourt et al., 2017 for a review).
Cumulatively, these factors suggest that cyberbullying may be more
complex and more insidious than traditional bullying. And yet, the
cyberbullying literature is still nascent with advancements in theory
and empirical evidence required to fuel developments in prevention
and intervention programming. A call-to-action is necessary given
several factors. First, findings from the Pew Research Center suggest
that 95% of teens own or have access to a smartphone and 45% of teens
surveyed report being “online on a near-constant basis” (Anderson &
Jiang, 2018). Twenty-four percent of youth surveyed by Pew felt that
social media had a negative impact on their lives; 27% of these youth
believe that social media is the reason for greater bullying and rumor
spreading (Anderson & Jiang, 2018). Second, mounting evidence exists
on the long-term negative outcomes uniquely associated with cyber-
bullying (Fahy et al., 2016; Vaillancourt et al., 2017) which is parti-
cularly troubling in light of the fact that roughly one in five adolescents
experience it (Hinduja & Patchin, n.d.; Selkie, Fales, & Moreno, 2016).
Vaillancourt et al. (2017) highlight that over and above traditional
bullying, cyberbullying is uniquely associated with (a) somatic diffi-
culties (e.g., headaches, stomach aches, etc.); (b) depression; (c) sui-
cidal ideation and suicide attempts; (d) symptoms of Post-Traumatic
Stress Disorder (PTSD); and (e) academic difficulties, among others.
Given the current state of the field, this literature review provides a
critical synthesis of the extant knowledge concerning (1) a definition of
cyberbullying; (2) theories explaining cyberbullying; (3) prevalence
rates; (4) a brief developmentally-focused overview of adolescents and
their online use; (5) risk and protective factors; (6) outcomes, over and
above traditional bullying; and (7) a brief overview of prevention and
intervention programming with information for key stakeholders. In
order to conduct this review, a literature search of the most current
scholarship published in peer-reviewed outlets from 2010 and beyond
was conducted. Keyword searches included cyberbullying, cybervictimi-
zation, electronic bullying, online bullying, and technology and bullying.
Only English language publications were used. Reports from large-scale

⁎
Corresponding author at: Rider University, Department of Psychology, 2083 Lawrenceville Road, Lawrenceville, New Jersey 08648-3099, United States of
America.
E-mail address: nansary@rider.edu.

https://doi.org/10.1016/j.avb.2019.101343
Received 9 October 2018; Received in revised form 9 September 2019; Accepted 15 October 2019
Available online 01 November 2019
1359-1789/ © 2019 Published by Elsevier Ltd.
N.S. Ansary
studies by government agencies and evidence-based advocacy groups
were also included. Furthermore, this review mainly focuses on ado-
lescents who are defined “as individuals in the 10–19 years age group”
(World Health Organization [WHO], 2013). Consistent with that defi-
nition, in this paper the term adolescent generally refers to middle- to
high-school-aged individuals who are distinct from college-aged sam-
ples.
In terms of the framework guiding this literature synthesis, devel-
opmental psychopathology provides a research-to-application con-
tinuum which is most useful in addressing this problem. Briefly, as a
discipline, developmental psychopathology explores the intersection of
human development (e.g., exploring change that recognizes the in-
dividual as an active agent in his or her development) and clinical
psychology (e.g., concerned with the diagnosis and treatment of dis-
ordered behavior) with a specific focus on using research to inform
prevention strategies (Hinshaw, 2013). Many scholars recognize that
developmentally-focused social-ecological frameworks that view bul-
lying as a dynamic process involving many levels of influence (e.g.,
family, school, neighborhood) are most useful in explaining bullying
(Swearer & Hymel, 2015; Zych et al., 2019). Furthermore, recent evi-
dence parsing risk and protective factors uniquely associated with cy-
berbullying are also helpful in conceptualizing the issue within a de-
velopmental psychopathology framework (Kowalski et al., 2019; Zych
et al., 2019). To date, no literature exists examining cyberbullying
through this lens, leaving a void in the scholarship and evidence-based
recommendations for best practices to address cyberbullying.
1. Operational definition
Failure to have a consensus about a definition is not a trivial pro-
blem. It is especially noteworthy that many researchers vary in how
they operationally define cyberbullying and this necessarily has im-
plications for the rigor of its study and for prevention programming.
How can adequate cyberbullying prevention strategies be effectively
designed when the definition and scope of the problem remain a
moving target?
There are two characteristics of cyberspace that present challenges
to the existing definition of traditional bullying which requires: an in-
tent to harm, a power imbalance, and repetition of the harmful beha-
vior (Olweus, 1999). First, many have argued that the power imbalance
criteria—either with regard to physical or social power required in the
traditional bullying definition—is less explicit in the online context.
Cyberspace inherently contributes to a power imbalance: the disin-
hibition effect affords online users opportunities to behave more bra-
zenly (Suler, 2004), and to attack with anonymity (Hinduja & Patchin,
2015a; see also Peter & Petermann, 2018). Another contributor to the
power imbalance is the fact that a single malicious act may be re-
peatedly disseminated and viewed by multiple users: This also negates
the intentional repetition criteria within the traditional bullying defi-
nition (Hinduja & Patchin, 2015a; see also Peter & Petermann, 2018).
Some researchers have attempted to sidestep this dilemma by
simply using global measures of cyberaggression (i.e., asking about
cybervictimization without any inquiry about repetition) as indicators
of cyberbullying (Kowalski et al., 2019; Peter & Petermann, 2018). The
implications of this are two-fold: operationalizing cyberbullying as sy-
nonymous with cyberaggression has compromised the internal validity
of those studies, and it has also obfuscated a clear understanding of
prevalence rates. For example, in their review of the literature, Hinduja
and Patchin (n.d.) found victimization rates that ranged from 0.4% to
92% and offending values ranged from 1.0% to 60.4%. Similarly, vast
prevalence rates have been reported in other large-scale reviews of the
cyberbullying literature (Kowalski et al., 2019; Selkie et al., 2016).
Beyond failure to make a distinction between cyberaggression and cy-
berbullying, other reasons for this vast range of prevalence rates in-
clude (1) the time parameter used in measurement (i.e., how many
times has this happened in the last 30 days or in the last year?, etc.); (2)
2
Aggression and Violent Behavior 50 (2020) 101343
age of participants (since cyberbullying during elementary school is
generally low but increases in adolescence; see Kowalski et al., 2019);
(3) whether a single item or multiple items were used (with multiple
item measures showing higher validity); and (4) cross-country com-
parisons where there may be differences in the availability of tech-
nology (Kowalski et al., 2019; Whittaker & Kowalski, 2015).
In an important contribution to the field, Peter and Petermann
(2018) distilled the essential elements of existing cyberbullying defi-
nitions in their concept analysis of articles published since 2010. Their
findings led to the following definition of cyberbullying as “using in-
formation and communication technologies (ICT) to repeatedly and
intentionally harm, harass, hurt and/or embarrass a target” (p. 359).
This definition is not significantly different than earlier forms which are
most often cited in the literature: Smith et al. (2008) define cyberbul-
lying as “an aggressive, intentional act carried out by a group or in-
dividual, using electronic forms of contact, repeatedly and over time
against a victim who cannot easily defend him or herself” (p. 376); and
Hinduja and Patchin (n.d.) define it as “… when someone repeatedly
and intentionally harasses, mistreats, or makes fun of another person
online or while using cell phones or other electronic devices”. Cumu-
latively, these nearly identical definitions reveal that a consensus has
developed.
2. Theoretical models
Similar to the lack of agreement about a cyberbullying definition,
there is no consensus about a comprehensive theoretical model that
explains cyberbullying. The largely a-theoretical state of the cyberbul-
lying literature has provided a barrier to a holistic understanding of the
issue (Kowalski, Giumetti, Schroeder, & Lattanner, 2014). Only one
theoretical model, to date—the Barlett and Gentile cyberbullying model
(BGCM, Barlett, 2017)—explicitly examines cyberbullying perpetra-
tion. Most scholars examining cyberbullying have adapted models of
aggression, such as the general aggression model (Anderson &
Bushman, 2002; see Barlett, 2017; Savage & Tokunaga, 2017), and I3 (I-
cubed) theory (Slotter & Finkel, 2011; see Wong, Cheung, & Xiao,
2018). Beyond explicit aggression models, the social-ecological dia-
thesis-stress model has been used to explain bullying (Swearer and
Hymel, 2015) and likely has great utility in explaining cyberbullying.
Each of these models will be briefly evaluated.
The Barlett and Gentile cyberbullying model (BGCM, Barlett, 2017)
accounts for (1) anonymity in cyberspace; (2) the recognition that one's
physical size does not contribute to a power imbalance online; and (3)
both of these mediate positive attitudes towards bullying, ultimately
predicting to cyberbullying perpetration. While the model has been
supported in examinations by its creators (Barlett, 2017; Barlett,
Madison, Heath, & DeWitt, 2019), there are several issues with its
usefulness. First, the theory is too simplistic given that it ignores prior
victimization as a catalyst for online perpetration of bullying which has
been found to be a significant correlate of cyberbullying (Kowalski
et al., 2014; Kowalski et al., 2019; Waasdorp & Bradshaw, 2015; see
also Wong et al., 2018). Second, person-specific factors (e.g., self-con-
trol, moral disengagement, use of technology, etc.) and context-specific
factors (e.g., parental factors including monitoring of technology,
schools failing to have explicit policies on consequences for cyber-
attacks, etc.) that have been found to be uniquely associated with cy-
berbullying perpetration (Kowalski et al., 2019; Zych et al., 2019), are
not represented in the model. Third, the operational definition of the
construct belief in the irrelevance of muscularity during online bullying (Bi-
MOB) has several issues, namely, (a) it is ambiguous, (b) careful con-
sideration of the items reveals they not only assess the belief that ‘size
does not matter’ online but concurrently assess the individual's positive
attitudes towards cyberbullying which is another unique predictor in
the model, and (c) the scale has shown varied reliability with reported
alpha levels ranging from as low as 0.52 to greater than 0.69 (Barlett
et al., 2019). Most importantly, the model has been examined only with
N.S. Ansary
adult samples ages 19–74 (Barlett et al., 2019), solely uses online data
collection procedures, and has been empirically examined only by its
creators.
Other models that have shown some utility in explaining cyber-
bullying are based on adapted versions of the generalized aggression
model (Anderson & Bushman, 2002) which examines person-specific
factors (e.g., demographic variables, personality, etc.) in confluence
with situation-specific factors (e.g., presence of bystanders, school cli-
mate, etc.) (Kowalski et al., 2014; Savage & Tokunaga, 2017). In ap-
plying this model, Savage and Tokunaga (2017) found that rather than
a direct relationship between person-specific predictors and cyberbul-
lying as an outcome, the relationship is much more complex and dy-
namic. In that study, high cyberbullying perpetration was demonstrated
among two sets of individuals (1) ones who were rated high on internet
self-efficacy, low on verbal aggressiveness, and high on social skills; and
(2) ones who were rated high on internet self-efficacy, high on verbal
aggressiveness, and low on social skills. These interactions suggest that
a confluence of many factors is at play in explaining cyberbullying since
no one factor universally predicted to the outcome. Moreover, the
findings suggest that extension of traditional bullying models to explain
cyberbullying are inadequate (Savage & Tokunaga, 2017).
I3 (I-cubed) theory (Slotter & Finkel, 2011), developed to explicate
intimate partner violence, has been adapted by Wong et al. (2018) to
examine cyberbullying perpetration. I3 theory examines pre-
dictors—both at the individual and situational level of—instigation (i.e.,
factors that provoke aggression), impellance (i.e., factors that promote
cyberbullying), and inhibition (factors that reduce the likelihood of cy-
berbullying perpetration). While the model demonstrated explanatory
power, it requires replication with younger samples in the West since
the sample under investigation was Hong Kong undergraduates (Wong
et al., 2018). Furthermore, the authors use a broader definition of cy-
beraggression (e.g., receiving messages that are harmful but with no
indication as to whether these happened repeatedly or not) and thus
replication with a more stringent measurement of cyberbullying is ne-
cessary.
In contrast to these models of aggression, Swearer and Hymel's
(2015) social-ecological diathesis-stress model provides a complex
framework that recognizes the dynamic interplay of genetics, social,
and environmental factors in explaining bullying perpetration. The
model accounts for how various settings (e.g., family, school, neigh-
borhood) in a child's life can interact with the child's genetic vulner-
abilities (e.g., temperament, personality, etc.), and individual risk and
protective factors to explain bullying involvement. Importantly, from a
developmental psychopathology framework, the model accounts for a
dynamic interplay of genetic vulnerability, environmental stressors,
and the norms of environmental contexts such as school, in explaining
involvement in bullying. Because of its complexity, the social-ecological
diathesis-stress model provides a way to integrate the empirical evi-
dence on the issue in a holistic model. Although the model was pro-
posed specifically for traditional bullying, it has power when con-
sidering cyberbullying. Consider the following factors, all represented
in this explanatory model, (a) a child's propensity to aggress (e.g., ge-
netic vulnerability, moral disengagement; etc.); (b) prior experiences
with being bullied (environmental stressors); (c) moderating factors
related to cyberspace (e.g., strength of the disinhibition effect and one's
technological efficacy); and (d) parental factors (e.g., parent-child re-
lationships and monitoring of technology which are context-specific
factors).
While the models noted here possess strengths in explaining cy-
berbullying perpetration, many require modification to sufficiently
capture its complexity. Adequate representation of personal, familial,
school, peer, and neighborhood/community contexts, as well as the
most current set of risk and protective factors examined in the literature
are lacking in the extant models. Accordingly, incorporating all of these
elements is critical in the refinement of existing models and in the
development of new ones.
3
Aggression and Violent Behavior 50 (2020) 101343
3. Prevalence
A common yet concerning issue in the cyberbullying literature is the
wide range of prevalence rates. This variability has been largely due to
discrepancies in operationally defining and measuring cyberbullying in
a way that discriminates it from cyberaggression. Nonetheless, large
scale studies and reviews place the average annual cybervictimization
rates between 14%–21%. The U.S. Department of Health and Human
Services, Centers for Disease Control report that 14.9% of high
schoolers reported being cyberbullied in the last year (Kann, McManus,
Harris, & et al., 2018). Hinduja and Patchin (n.d.) conducted an ex-
tensive review of the literature (evaluating over 234 scholarly pub-
lications) and found a wide range of rates with the average lifetime
incidence rate of cybervictimization at 21% and 13% for cyberbullying
perpetration. These values are generally consistent with other data on
the incidence of cyberbullying (Kowalski et al., 2019; Olweus & Limber,
2018; Selkie et al., 2016).
4. Adolescents and apps
The pace of emerging technology that draws in a young population
of users creates challenges for the scholarly study of cyberbullying.
Social media applications (Apps) create increasingly sophisticated yet
easy-to-use modalities for youth to connect to one another in both
helpful and harmful ways. Suicides of young adolescents as a result of
cyberbullying have fueled a more urgent dialogue concerning the role
of social media Apps–such as Ask.fm and Kik–in cyberbullying (Smith-
Spark, 2013; Wallace, 2015). These—as well as other free Apps such as
Voxer, Whisper, Wut, Rumr—allow users to anonymously post photos
or messages that can be viewed for a short time online. Others like
Snapchat allow users to post publicly or privately to subsets of friends.
Likewise, fake accounts on Facebook, Instagram, and Twitter also afford
the opportunity to anonymously engage in cyberbullying perpetration.
Although, most social media platforms require a minimum user age of
13, many Apps are used by younger youth (see Valencia, 2014).
Emerging evidence suggests that there are age differences in the
ways that individuals engage social media to cyberbully. For elemen-
tary school-aged youth, online gaming (which affords users the ability
to converse via headset as well as in-game text) has been identified as
the most widespread way to engage in cyberbullying (see Kowalski
et al., 2019). Among adolescents, cyberbullying most commonly occurs
via social media (Kowalski et al., 2019) with one study identifying
Twitter and Facebook as the most commonly used social media plat-
forms to do so (Whittaker & Kowalski, 2015). Within a small young
adult sample, both public and private comments, status updates, and
posts, were the most common ways of engaging in cyberbullying (Brody
& Vangelisti, 2017).
School-aged social media users are an at-risk population for cyber-
bullying involvement. First, school-aged youth have nascent emotional
and cognitive regulatory systems in the brain, which govern behavior.
More specifically, the prefrontal cortex of the brain—which administers
impulse control, judgment, and thinking through anticipated con-
sequences of one's behavior—does not conclude development until
young adulthood (Pharo, Sim, Graham, Gross, & Hayne, 2011). Second,
some features of cyberspace—anonymity, ubiquity of technology, and
short time-duration of posts—afford opportunities for youth with an
intention to harm others to do so with deleterious consequences for the
target (Peter & Petermann, 2018; Wong et al., 2018; Zych et al., 2019).
To illustrate, the individual who bullies may be a complete stranger or
someone the target knows who is posting anonymously (or alternatively
using an alias or different identity) which can contribute to the target
feeling helpless. Moreover, the attacks may be occurring privately or
publicly and the target may receive posts that are viewable for an ex-
tremely short time (just a few seconds) or for much longer periods of
time as well as re-posted easily. Thus, cyberbullying incidents can be
privy to a potentially large online audience of peers and strangers
N.S. Ansary
leaving targets unsure of who has “seen” the incident (e.g., photo, ob-
scene posting, etc.). Third, the frequency of cyberbullying intervention
is quite low: Adults are not typically present online and evidence sug-
gests that nearly 91% of bystanders report that they did nothing when
they witnessed such an attack (Lenhart et al., 2011). Fourth, reporting
of cyberbullying incidence is very low since targets may fear parents
revoking their access to technology in order to protect them from fur-
ther abuse. The confluence of low adult supervision, the perception that
cyberbullying will not be reported, low bystander intervention, and the
lack of repercussions for such behavior are all potent contributors to
cyberbullying perpetration (Kowalski et al., 2019; Zych et al., 2019).
Cumulatively, the factors presented in this section provide the
backdrop for cyberbullying involvement. The convergence of these
factors sets the stage for targets experiencing helplessness, psycholo-
gical distress, as well as fear that there is no safe place online (Dehue,
2013; Vandebosch & Van Cleemput, 2009). From this perspective, it is
clear why cyberbullying involvement is associated with long-term
psycho-social dysfunction, above and beyond the effects seen for tra-
ditional bullying.
5. Risk and protective factors
Recent years have seen an increase in the identification of risk and
protective factors of cyberbullying, and of course, many of these are
shared with traditional bullying. Two reviews on this issue, Kowalski
et al. (2019) and Zych et al. (2019), are particularly rigorous in their
focus on the unique risk and protective factors of cyberbullying. The
realms of risk and protective factors identified in those reviews relate to
the domains seen in Swearer & Hymel's social-ecological diathesis-stress
model (Swearer and Hymel, 2015), namely, individual, family, peer,
school, and community. Before proceeding through the list of risk and
protective factors, it is important to point out that in a review of meta-
analyses considering protective factors on this issue (Zych et al., 2019),
the protective factors for cybervictimization with the strongest effect
size were those at the individual level related to personal competencies
and low technology use, while for cyberbullying perpetration, the
strongest protective effects were found for low technology use and peer-
level factors. It is also critical to note that no prospective longitudinal
studies have been conducted to identify risk and protective factors of
cyberbullying, and existing findings are mainly garnered from cross-
sectional designs (Zych et al., 2019).
For cyberbullying victimization, individual-level risk factors in-
clude: low self-esteem, self-control, social intelligence, low empathy,
high levels of anxiety, aggression, moral disengagement, and being a
victim of traditional bullying victimization (Kowalski et al., 2019).
Stress and suicidal ideation have also been identified as risks for cy-
bervictimization (Kowalski et al., 2014). Protective factors of cyber-
bullying victimization include high self-esteem (Kowalski et al., 2019)
as well as high levels of emotional management, self-concept, and social
competence (Zych et al., 2019). With regard to technology use, low
levels have been found to be protective (Zych et al., 2019) while high
levels are a risk (Kowalski et al., 2019). In the parenting domain, risks
include a single-family household, negative family environment, and
sexual abuse (Kowalski et al., 2019), while protective factors consist of
parental support and warmth (Kowalski et al., 2019) and high socio-
economic status (Zych et al., 2019). Just as low peer support has been
identified as a risk for cybervictimization (Kowalski et al., 2019), high
peer support and positive peer interactions have been identified as a
protective influence (Zych et al., 2019). School and community factors
also play a role in that school climate, safety, satisfaction with school,
and a safe neighborhood have been identified in these two reviews as
protective factors against cyberbullying victimization (Kowalski et al.,
2019; Zych et al., 2019).
Regarding group differences, there is an emerging trend in terms of
gender differences and there is largely mixed evidence when con-
sidering disparities by race and ethnicity. With respect to gender
4
Aggression and Violent Behavior 50 (2020) 101343
differences, girls appear to be at greater risk (Kann et al., 2018;
Kowalski et al., 2019), and this will be addressed in more detail in the
next section. Explorations into ethnic and racial differences in cyber-
victimization have yielded mixed-results (Kowalski et al., 2019),
though some evidence suggests that whites experience cybervictimi-
zation at higher rates compared to their black and Hispanic counter-
parts (Kann et al., 2018). Additionally, there is some evidence that
Arab-Americans may be distinctly targeted in cyberbullying (Kowalski
et al., 2019). Other forms of bias-based bullying are supported in the
literature in that the LGBTQI community, those with disabilities, and
individuals with elevated weight status have also been identified as at-
risk (Kowalski et al., 2019). The Youth Behavior Risk Surveillance from
the Centers for Disease Control found that LGBTQI youth reported cy-
bervictimization rates over twice as high as those reported by their
heterosexual counterparts (27.1% and 13.3% respectively) (Kann et al.,
2018). Absent from the literature is any mention of religious minorities.
Factors predictive of cyberbullying perpetration have also been
documented in the literature. For cyberbullying perpetration, in-
dividual-level risk factors include: low self-esteem, empathy, self-con-
trol, and high levels of impulsivity, moral disengagement, anxiety, ag-
gression, and traditional bullying victimization (Kowalski et al., 2019).
Alternatively, high self-esteem, social competence, and emotional
management have been identified as protective factors (Kowalski et al.,
2019; Zych et al., 2019). High technology use has been identified as an
individual-level risk factor for cyberbullying perpetration (Kowalski
et al., 2019), and low use has been identified as a protective factor
(Zych et al., 2019). With regard to the family realm, a positive parent
relationship and home environment have been found to be protective
(Kowalski et al., 2019; Zych et al., 2019), and parental supervision of
technology has also been identified as a protective factor (Zych et al.,
2019). Regarding the peer domain, positive peer interactions as well as
peer support have been identified as protective factors (Kowalski et al.,
2019; Zych et al., 2019). Conversely, peer rejection, lack of prosocial
peers and peer norms supporting bullying have been identified as risks
(Kowalski et al., 2019). School and community factors including a po-
sitive school climate and satisfaction with school have found to be
protective against cyberbullying perpetration (Kowalski et al., 2019;
Zych et al., 2019). With regard to group differences, the evidence on
gender differences and race and ethnic disparities in cyberbullying is
mixed (Kowalski et al., 2019).
There is substantial overlap in the risk and protective factors for
both victims and perpetrators of cyberbullying. This is likely due to the
most vulnerable group of youth who are both targets of bullying and
perpetrators of it—bully-victims. More work is necessary to better un-
derstand this vulnerable group in order to design more effective pro-
gramming to address their needs.
5.1. Gender differences
Mixed evidence exists about gender differences in cyberbullying
though a preponderance of evidence points to females being at greater
risk of involvement (Kann et al., 2018; Kowalski et al., 2019). In point
of fact, the most current results of the Youth Behavior Risk Surveillance
from the Centers for Disease Control indicate that females reported
nearly twice as high cybervictimization as males (19.7% and 9.9% re-
spectively) (Kann et al., 2018). Other studies suggest no gender dif-
ferences (see Wong et al., 2018 for a brief review).
Such inconsistencies in the literature point to the likelihood that
that there are more nuanced differences in the ways that males and
females relate online and how each group interacts in cyberbullying
situations. According to Wong et al. (2018), males in their study were
more likely to engage in cyberbullying perpetration if they had been
previously victimized, whereas for females, they were more motivated
by the disinhibition effect. These gender differences imply that males
and females engage in cyberaggression for different reasons in cyber-
space: males possibly using it to seek revenge while females may feel
N.S. Ansary
emboldened in situations where their accountability is diminished on-
line. The motivation for males to seek revenge for prior cyberbullying
concurs with the overrepresentation of males in the cyberbully-victim
category found in other studies (Fahy et al., 2016).
Beyond gender differences in motivations for cyberbullying perpe-
tration, the literature has also begun to explore gender differences in
the topics that individuals who perpetrate cyberbullying choose to use.
For example, one study found that in a sample of undergraduates, fe-
males were more likely to be targeted with topics about sexual activity
(e.g., nude photos, etc.), whereas men were more frequently cyber-
bullied about their skills/talents (e.g., intelligence) and sexual or-
ientation (Brody & Vangelisti, 2017). Again, these findings not only
point to the need for further work exploring gender differences but also
should catalyze more in-depth examination of bias-based forms of cy-
berbullying.
6. Outcomes
Vaillancourt et al. (2017) provide an excellent overview of the un-
ique effects of cyberbullying—over and above traditional bullying—on
mental wellbeing. The plethora of research reviewed by them indicate
that after controlling for traditional bullying, cyberbullying is asso-
ciated with: anxiety, depression, suicidal ideation and suicide attempts,
somatic complaints and compromised physical health, symptoms of
post-traumatic stress disorder (PTSD), and academic difficulties. Re-
garding intensity of psychopathology, the authors state “a dose-re-
sponse effect is commonly found between being cyberbullied and the
severity of its consequences—youth who are bullied the most are the
ones who suffer the most” (Vaillancourt et al., 2017, p. 370).
Similar conclusions have been drawn by others about the ill-effects
of cyberbullying. In a longitudinal study which controlled for baseline
levels of distress, cyberbullying was associated with higher social an-
xiety, depression, and lower overall wellbeing at one-year follow-up
(Fahy et al., 2016). In that study, no long-term effects were found
among individuals who engaged in cyberbullying perpetration, how-
ever the study was limited in that it did not include externalizing be-
havior dimensions for consideration. Likewise, in a study only con-
sidering negative experiences on Facebook, participants who reported
any form of negative interaction, regardless of recency of the incident,
showed elevated depression levels after controlling for baseline de-
pression (Rosenthal, Buka, Marshall, Carey, & Clark, 2016). With regard
to suicidal ideation, while both traditional bullying and cyberbullying
were found to be associated with higher suicidal ideation, youth who
reported being a victim of both forms were over five times as likely to
report experiencing suicidal ideation (Hinduja and Patchin, 2019a, b).
7. Prevention and intervention
As a discipline, developmental psychopathology requires specific
criteria in the design of effective prevention and intervention pro-
gramming, including (1) theory which provides a foundation from
which all strategies are based on; (2) an emphasis on risk and protective
factors that have been identified in the empirical literature; (3) pro-
minence of the various contexts that impact an individual (i.e., home,
school, etc.) in program design; and (4) empirical evaluation of the
program's effectiveness (Hinshaw, 2013). Accordingly, there are several
limitations in the programming literature on the issue, namely (1)
programs have either been designed as extensions of traditional bul-
lying programs, or a-theoretically (Tanrikulu, 2018); (2) much of the
literature focuses on examining factors influencing involvement in cy-
berbullying on a one-context-at-a-time basis (i.e., parental monitoring
or school-based remedies), but seldom together in a holistic framework;
and (3) program evaluation has been limited due to variations in
“duration, theory, sampling, design, and measurement tools”
(Tanrikulu, 2018, p. 84).
Prior to addressing the various contexts for which programming has
5
Aggression and Violent Behavior 50 (2020) 101343
been developed, it is important to discuss the findings of the only
published cyberbullying program reviews to date (1) Van Cleemput
et al. (2014) reviewed eight programs across nine countries and con-
ducted a meta-analysis; (2) Tanrikulu (2018) reviewed seventeen pro-
grams also across the same nine countries; and (3) Gaffney, Farrington,
Espelage, and Ttofi (2019) included twenty-four publications published
from 2012 onward in their meta-analysis on cyberbullying program
effectiveness. All three analyses examined studies based on school-aged
samples and all concluded that programs demonstrated only modest
effects. The meta-analysis conducted by Van Cleemput et al. (2014)
found an average effect size for reducing victimization g = 0.135
(n = 9453, k = 6, 95% CI 0.079; 0.190, p < .001), while for perpe-
tration it was g = 0.065 (n = 6373, k = 6, 95% CI 0.019; 0.112,
p < .01). Alternatively, Gaffney et al. (2019) found overall reductions
in cyberbullying victimization to range between 14 and 15% and for
cyberbullying perpetration to range from 9 to 15%. These rates are
modest and are commensurate with the effects gleaned in traditional
bullying programs (Gaffney et al., 2019; Van Cleemput et al., 2014).
Regarding program strategies at school, Gaffney et al. (2019) did not
explicitly analyze this and Van Cleemput et al. (2014) concluded that
most programs did not adopt a whole school approach, but rather fo-
cused on modules implemented to address cyberbullying added on to
their traditional anti-bullying programs. This is unfortunate given the
large body of literature noting the centrality of a whole school approach
in bullying prevention (see Ansary, Elias, Greene, & Green, 2015a,
2015b for a review).
Given the dearth of holistic programs addressing cyberbullying, I
examine program strategies that are sensitive to risk and protective
factors and that are also context-specific. This perspective aligns with
Swearer and Hymel's (2015) social-ecological diathesis-stress model,
which conceptualizes the individual who bullies in a dynamic interac-
tion with his/her home, school, peer-network, and neighborhood/
community. Thus, the following section will examine evidence-based
program strategies for parents, schools, bystanders, and social media
platforms separately.
7.1. Parents
Positive parenting dimensions have been found to be a protective
factor for both cybervictimization and cyberbullying perpetration
(Kowalski et al., 2019). Parental supervision of technology is also a
protective factor in cyberbullying perpetration (Zych et al., 2019). This
is no surprise as parents are recognized as an important target audience
in traditional bullying prevention (Ansary et al., 2015a; Ansary et al.,
2015b), and they may be even more critical agents in cyberbullying
prevention and intervention.
With regard to monitoring of technology, according to Pew
Research Center (Anderson, 2016), parents may be relatively hands-on
in monitoring their children's technology use. In point of fact, 60% of
parents say they check their teens' social media profiles, and 48% have
ever looked through their teens' phone calls or texts (Anderson, 2016).
And while many parents recognize this important role, they report
lacking the technological knowledge to comprehensively monitor
(Kessel Schneider, Smith, & O'Donnell, 2013). Although online tip
sheets are an important source of information for parents, Espelage and
Hong (2017) caution that these may not be vetted and accordingly
recommend that parents seek guidance from government agencies or
advocacy organizations that are evidence-based.
Guidance for parents should include information on prevention
(e.g., teaching children digital citizenship) and intervention when cy-
berbullying happens (e.g., block users who post harmful messages).
Regarding prevention, parents should instruct their children about safe
ICT use, such as (1) not sharing usernames or passwords; (2) not pro-
viding personal information in profiles, chat rooms, and other fora; (3)
not responding to threatening messages and to notify an adult im-
mediately; and (4) shutting technology off if a threatening message is
N.S. Ansary
received (National Crime Prevention Council, n.d.). Nurturing re-
sponsible use of technology also entails instructing youth (1) about
refraining from sending personal or inappropriate photos of oneself; (2)
that a post leaves a permanent footprint that cannot be taken back
(National Crime Prevention Council, n.d.); (3) to tell an adult without
fear of losing access to their technology; and (4) that consequences will
be firmly enforced for engaging in cyberbullying perpetration (see
Hinduja & Patchin, 2013). With respect to intervention when cyberbul-
lying occurs, evidence suggests that engaging the child who bullies in
further online communication about the incident exacerbates the pro-
blem (Perren et al., 2012), and accordingly parents should immediately
block the potential cyberbully from any further communication. Re-
porting the attack to the appropriate social media outlet is also a critical
step. Although parents may be challenged to effectively monitor chil-
dren's ICT use, they can certainly make a difference.
7.2. Educators and schools
School climate and satisfaction with school are protective factors for
both targets of cyberbullying and individuals who perpetrate cyber-
bullying (Kowalski et al., 2019; Zych et al., 2019). In prior research,
cyberbullying was characterized as occurring outside of school with
negative interactions continuing into the next school day (see Cassidy,
Faucher, & Jackson, 2013). More recent trends indicate that the di-
chotomy between technology use at school versus at home is no longer
an important distinction. Adolescents often have access to computers or
tablets while at school, and though many schools ban cell phone use,
the reality is that most youth use them during the school day. In a
similar vein, schools also provide the context within which peers in-
teract socially and since most targets of cyberbullying know the per-
petrator (Whittaker & Kowalski, 2015), schools must play an important
role in prevention and intervention. Cumulatively, these risks under-
score the need for effective and explicit school-based cyberbullying
programming. As will be discussed later, laws in all fifty states in the
U.S. hold schools accountable for preventing, intervening, as well as
investigating and reporting cyberbullying (U.S. Department of Health
and Human Services, 2018).
Fostering a positive school climate is crucial in the prevention of
traditional bullying and cyberbullying (Ansary et al., 2015a; Ansary
et al., 2015b; Espelage & Hong, 2017). Moreover, Espelage and Hong
(2017) note that with regard to cyberbullying, schools should (a) have
clear proactive policies, procedures, and practices about internet use
and cyberbullying; (b) actively promote awareness of social-emotional
skills among faculty, staff, and students; (c) promote a positive school
environment; and (d) nurture school-family connections. While these
are no different than essential elements of effective school-based anti-
bullying programs, an explicit emphasis on bullying in cyberspace is
critical. Some evidence suggests that cyberbullying is not always ex-
plicitly mentioned in schools' anti-bullying policies (Kessel Schneider
et al., 2013). This is unfortunate in light of research evidence suggesting
that youth who report having clear rules about cyberbullying from
school and parents are significantly less likely to report involvement in
online bullying (Hinduja & Patchin, 2013).
Other areas where schools can work towards preventing cyberbul-
lying, is to directly tackle the issue through the school's curriculum in a
developmentally appropriate fashion. The evidence is clear that social
emotional character development (SECD) is one of the central columns
that effective anti-bullying programs are built on (Ansary et al., 2015a;
Ansary et al., 2015b). The cyberbullying prevention literature is rife
with synonyms for socially responsible behavior (e.g., prosocial beha-
vior, kindness, etc.) though many researchers do not specifically refer to
these as social-emotional skills. Just as it is in traditional bullying
prevention, SECD should play a prominent role in cyberbullying pre-
vention. Another critical element in addressing cyberbullying in schools
is professional development: Teachers need to be trained to identify,
respond, and report incidents to the necessary channels the school has
6
Aggression and Violent Behavior 50 (2020) 101343
in place to address cyberbullying (Ansary et al., 2015a; Ansary et al.,
2015b; Espelage & Hong, 2017).
7.3. Peers and bystanders
For both cyberbullying targets and perpetrators, positive peer in-
teractions and peer support are protective factors (Kowalski et al.,
2019; Zych et al., 2019). Conversely, peer rejection, lack of prosocial
peers and peer norms supporting bullying are risks for cyberbullying
perpetration (Kowalski et al., 2019). Allison and Bussey (2016) found
that one of the strongest predictors of cyberbullying bystander inter-
vention is whether or not the target of the bullying is a friend.
Bystanders play a key role in defusing or escalating incidents in both
traditional and cyberbullying domains (Allison & Bussey, 2016). Ac-
cording to Lenhart et al. (2011) 88% of U.S. teens have witnessed cy-
berbullying and of these, 91% reported that they did nothing about it.
Lack of bystander intervention can be explained by several factors,
namely (a) whether posts occur on large platforms where youth feel
that others will likely intervene (diffusion of responsibility); (b) worry
over what others will say; and (c) because of the asynchronous nature of
cyber-interactions, youth may feel that it is too late to act (Allison &
Bussey, 2016). According to Allison and Bussey (2016) these char-
acteristics of bystander intervention have been replicated cross-cultu-
rally.
7.4. Social media
Social media companies must walk the fine line between regulating
free speech (i.e., policing and removing content that may be harmful or
offensive), and protecting the rights of users (Milosevic, 2016). Con-
sequently, these companies often “…downplay their own responsibility
for the extent to which technologies may facilitate such behavior…”
and “they see users as better positioned to understand the context be-
hind their own conflicts” (p. 5175). Accordingly, many social media
companies advocate for social reporting or community moderation to
address cyberbullying because it is more efficient for the company
though not necessarily in the best interest of users (Milosevic, 2016).
Larger social media companies tend to have a more comprehensive
definition of cyberbullying as well as more sophisticated policies and
procedures in place to help support users compared to younger com-
panies (Milosevic, 2016). Most social media platforms have mechan-
isms to prevent and intervene in cyberbullying such as: reporting,
blocking, filtering software, as well as human and automated detection
systems. Automatic detection through feature engineering (Rosa et al.,
2019) examines various characteristics of posts (e.g., language, emoti-
cons, etc.) to identify instances of cyberbullying. In a review of auto-
matic cyberbullying detection studies, Rosa et al. (2019) conclude that
current detection systems do not distinguish between cyberaggression
and cyberbullying: Those researchers recommend annotator training on
the proper definition and overall concept of cyberbullying (Rosa et al.,
2019). Further, automatic detection systems may be less efficient since
human moderators can understand nuances in language such as irony
or context-specific aggression (Milosevic, 2016).
Beyond explicit guidance on intervention, social media companies
should do more to prevent cyberbullying. Some prevention efforts em-
ployed by social media platforms include (1) messaging about appro-
priate online behavior and a focus on positive instead of negative in-
teractions (Barlett, DeWitt, Maronna, & Johnson, 2018; Kessel
Schneider et al., 2013); (2) notifications and delays that are designed to
promote self-reflection and prosocial behaviors (Rosa et al., 2019), such
as proactive screening, which can identify potentially harmful posts and
prompt the user with messages such as “Are you sure you want to post
this”? or “Say something kind” (see Milosevic, 2016, p. 5172); (3) re-
quire users' to reveal their identities since sites that do so report lower
levels of cyberbullying than those that do not (Barlett et al., 2018); and
(4) require users to authenticate their account and emails regularly to
N.S. Ansary
prevent the use of fake accounts (Barlett et al., 2018). Furthermore,
social media platforms should encourage bystanders to intervene
(Barlett et al., 2018) even if by the more commonly seen indirect
methods of intervention (e.g., rating a post unfavorably or reporting the
perpetrator) as opposed to direct intervention (explicit correspondence
about the incident) (Barlett et al., 2018).
7.5. Health care providers
Since reporting of cyberbullying involvement is extremely low
(Cassidy et al., 2013), medical doctors and mental health practitioners
working with adolescents may play important roles in detecting it.
These health professionals can ask questions about cyberbullying (ei-
ther directly or by survey) at intake and encourage youth to speak with
parents about limiting technology (Espelage & Hong, 2017). For mental
health practitioners, the link between cybervictimization and myriad
forms of long-term psychopathology are well documented (Vaillancourt
et al., 2017). Furthermore, the important role of mental health practi-
tioners is underscored by evidence (that requires replication) sug-
gesting that even incidents that happened in the past are associated
with present-day dysfunction (Rosenthal et al., 2016).
7.6. Law
Although all fifty states in the U.S. have laws addressing cyberbul-
lying, there is considerable variability in their protections and the U.S.
Department of Education has provided federal guidance on the issue
since 2010, in an effort to promulgate a basic standard for protections
(U.S. Department of Health and Human Services, 2018). Given varia-
bility across states, Hinduja and Patchin, (2015b) outline six re-
commendations for laws addressing effective anti-cyberbullying po-
licies. These include (1) a specific definition of harassment,
intimidation, and bullying with explicit mention of various technolo-
gies; (2) sanctions that increase in severity based on repetition; (3) clear
policies for reporting; (4) clear policies on investigating; (5) explicit
mention of activities occurring off-school grounds that constitute cy-
berbullying; and (6) policies and procedures for preventing cyberbul-
lying. These are nearly identical to best-practices for school-based
programming to address traditional bullying (Ansary et al., 2015a,
2015b). These laws hold educators and administrators responsible for
effectively intervening, investigating, and reporting, as well as for the
implementation of effective cyberbullying prevention programming.
8. Limitations and future directions
While it is encouraging to see an explosion in the empirical in-
vestigation of cyberbullying, significant limitations in the field persist
and more scholarship is required. The most general limitations in the
existing literature surround the difficulties of a definition, inconsistent
ways of measuring cyberbullying, and limited theories holistically ex-
plaining the issue. While progress has occurred in terms of a definition,
greater work is needed on the construction of reliable and valid mea-
surement of cyberbullying that effectively discriminates it from cyber-
aggression. Doing so will foment progress in terms of strengthening the
internal validity of studies examining cyberbullying, as well as in es-
tablishing accurate prevalence rates.
Regarding theoretical models explaining cyberbullying from a psy-
chological perspective, with the exception of Swearer and Hymel's
(2015) social-ecological diathesis-stress model, the existing models are
too simplistic and do not adequately account for the existing risk and
protective factors that have been empirically identified. Moreover, in
light of recent evidence suggesting gender differences in cyberbullying
perpetration (Kowalski et al., 2019; Wong et al., 2018), theoretical
models will need to explicitly account for gender. Inclusion of variables
assessing other minority identities (i.e., racial, ethnic, religious, sexual,
etc.) is also necessary as these individuals are at elevated risk of being
7
Aggression and Violent Behavior 50 (2020) 101343
targeted. Additionally, prior victimization is a critical variable that
should be present in theoretical models of cyberbullying, given that it
has been repeatedly identified as a risk factor (Kowalski et al., 2014,
2019; Waasdorp & Bradshaw, 2015; see also Wong et al., 2018). Such
insights necessitate the refinement of existing models to account for
these complexities and also point to the need for the creation of more
complex models that comprehensively capture the problem. Progress in
the development of holistic theoretical models that are tested among
diverse samples (e.g., heterogeneity in age, race, ethnicity, sexual or-
ientation, disability, etc.) and using various methodologies is needed to
undergird progress in the scholarly and applied realms.
There are no prospective longitudinal studies published regarding
risk and protective factors and thus the findings are mainly related to
cross-sectional research (Zych et al., 2019). Furthermore, more elabo-
rate research designs are needed to adequately capture the dynamic
interactions between individual, familial, peer, school, and community
levels in the exploration of risk and protective factors. For example,
Savage and Tokunaga (2017) concluded that cyberbullying perpetra-
tion was explained by a complex relationship of social skills, internet
self-efficacy, and verbal aggressiveness. As such, prospective long-
itudinal analyses that examine these dynamic interactions is necessary
and path analysis, structural equation modeling, and person-based
analysis will provide useful insights into the complexity of cyberbul-
lying.
Study of group differences in the risks for cyberbullying victimiza-
tion and perpetration has just begun. More work is required to further
explain the mixed evidence about gender differences, which can only
occur by exploring the differences in motivations for technology use,
the salience of the disinhibition effect, coping with prior victimization,
among other factors between males and females (Wong et al., 2018).
Moreover, scant evidence on bias-based forms of cyberbullying exist
(Kann et al., 2018; Kowalski et al., 2019) and these suggest groups of
understudied populations that are particularly vulnerable. In spite of
evidence regarding disproportionately higher rates of traditional bul-
lying of Muslim, Jewish, and Sikh youth (Ansary, 2018), there are no
peer-reviewed studies addressing cyberbullying of religious minorities.
While it may not be the case that these youth are explicitly targeted
with reference to their minority identity (e.g., Muslim youth being
called a terrorist), they are likely to be, more broadly, at elevated risk
for all forms of cyberbullying.
Evidence-based prevention and intervention efforts will benefit
from work that addresses the existing limitations in the field. At this
time, cyberbullying programs have limited or no theoretical basis, are
often considered as an add-on module to traditional bullying programs,
and have not been evaluated in meaningful ways to determine effec-
tiveness (Tanrikulu, 2018; Van Cleemput et al., 2014). Strides are ne-
cessary to build-on the growing body of research pointing to specific
risk and protective factors that are context-specific in order to design
more effective programs. Furthermore, there should be an urgent call
for policy makers to develop restrictions on social media applications
that are advertised to younger users that will ensure better prevention
and intervention of cyberbullying.
9. Conclusion
If the next ten years of research mirrors the exponential growth in
the cyberbullying literature that we have seen in the past ten years, the
field should be equipped with well-designed effective prevention and
intervention programming. Moreover, progress must address the em-
pirical evaluation and tailoring of prevention and intervention pro-
grams to address the needs of our most vulnerable youth, namely, ra-
cial/ethnic and religious minorities, LGBTQI individuals, and persons
with varying abilities. Empirically informed prevention strategies that
address the gamut of an adolescent's life will eventually offer the pro-
tection that every child deserves.
N.S. Ansary
Acknowledgements
I am deeply grateful to Alanna D'Avanzo and Syrah Liles for their
valuable feedback which have improved the manuscript greatly.
References
Allison, K. R., & Bussey, K. (2016). Cyber-bystanding in context: A review of the literature
on witnesses’ responses to cyberbullying. Children and Youth Services Review, 65,
183–194. https://doi.org/10.1016/j.childyouth.2016.03.026.
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of
Psychology, 53, 27–51. https://doi.org/10.1146/annurev.psych.53.100901.135231.
Anderson, M., & Jiang, J. (2018, May 31). Teens, social media & technology 2018.
Retrieved on May 19, 2019 from https://www.pewinternet.org/2018/05/31/teens-
social-media-technology-2018/.
Anderson, M. A. (2016). Parents, teens and digital monitoring. Washington, D.C: Pew
Research Center. January 7. Retrieved on June 14, 2019 from https://www.
pewinternet.org/2016/01/07/parents-teens-and-digital-monitoring/.
Ansary, N. S. (2018). Religious-based bullying: Insights on research and evidence-based best
practices from the National Interfaith Anti-Bullying Summit (2017). Washington, D.C:
Institute for Social Policy and Understanding. Retrieved on June 14, 2019 from
https://www.ispu.org/religious-based-bullying-insights-on-research-and-evidence-
based-best-practices-from-the-national-interfaith-anti-bullying-summit/.
Ansary, N. S., Elias, M., Greene, M., & Green, S. (2015a). Guidance for schools selecting
antibullying approaches: Translating evidence-based strategies to contemporary im-
plementation realities. Educational Researcher, 44(1), 27–36. https://doi.org/10.
3102/0013189X14567534.
Ansary, N. S., Elias, M., Greene, M., & Green, S. (2015b). Best practices to address (or
reduce) bullying in schools. Phi Delta Kappan, 97, 30–35. https://doi.org/10.1177/
0031721715610088.
Barlett, C. P. (2017). From theory to practice: Cyberbullying theory and its application to
intervention. Computers in Human Behavior, 72, 269–275. https://doi.org/10.1016/j.
chb.2017.02.060.
Barlett, C. P., DeWitt, C.C., Maronna, B., & Johnson, K. (2018). Social media use as a tool
to facilitate or reduce cyberbullying perpetration: A review focusing on anonymous
and nonanymous social media platforms. Violence and Gender, 5(3), 147–152. doi.
org/https://doi.org/10.1089/vio.2017.0057.
Barlett, C. P., Madison, C. S., Heath, J. B., & DeWitt, C. C. (2019). Please browse re-
sponsibly: A correlational examination of technology access and time spent online in
the Barlett Gentile Cyberbullying model. Computers in Human Behavior, 92, 250–255.
https://doi.org/10.1016/j.chb.2018.11.013.
Brody, N., & Vangelisti, A. L. (2017). Cyberbullying: Topics, strategies, and sex differ-
ences. Computers in Human Behavior, 75, 739–748. https://doi.org/10.1016/j.chb.
2017.06.020.
Cassidy, W., Faucher, C., & Jackson, M. (2013). Cyberbullying among youth: A com-
prehensive review of current international research and its implications and appli-
cation to policy and practice. School Psychology International, 34(6), 575–612. https://
doi.org/10.1177/0143034313479697.
Dehue, F. (2013). Cyberbullying research: New perspectives and alternative methodolo-
gies. Introduction to the special issue. Journal of Community & Applied Psychology, 23,
1–6. https://doi.org/10.1002/casp.2139.
Espelage, D. L., & Hong, J. S. (2017). Cyberbullying prevention and intervention efforts:
Current knowledge and future directions. The Canadian Journal of Psychiatry, 62(6),
374–380. https://doi.org/10.1177/0706743716684793.
Fahy, A., Stansfeld, S. A., Smuk, M., Smith, N., Cummins, S., & Clark, S. (2016).
Longitudinal associations between cyberbullying involvement and adolescent mental
health. Journal of Adolescent Health, 59(5), 502–509. https://doi.org/10.1016/j.
jadohealth.2016.06.006.
Gaffney, H., Farrington, D. P., Espelage, D. L., & Ttofi, M. M. (2019). Are cyberbullying
intervention and prevention programs effective? A systematic and meta-analytical
review. Aggression and Violent Behavior, 45, 134–153. https://doi.org/10.1016/j.avb.
2018.07.002.
Hinduja, S., & Patchin, J. W Cyberbullying Facts. Cyberbullying Research Center.
Retrieved on June 14, 2019 from https://cyberbullying.org/facts.
Hinduja, S., & Patchin, J. W. (2013). Social influences on cyberbullying behaviors among
middle and high school students. Journal of Youth & Adolescence, 42, 711–722.
https://doi.org/10.1007/s10964-012-9902-4.
Hinduja, S., & Patchin, J. W. (2015a). Bullying beyond the schoolyard: Preventing and re-
sponding to Cyberbullying (2nd ed.). Thousand Oaks, CA: Sage Publications.
Hinduja, S., & Patchin, J. W. (2015b). Cyberbullying legislation and case law:
Implications for school policy and practice. Retrieved on June 14, 2019 from
Cyberbullying Research Centerhttps://cyberbullying.org/cyberbullying-legal-issues.
pdf.
Hinduja, S., & Patchin, J. W. (2019a). Cyberbullying facts. Retrieved on June 14, 2019
from Cyberbullying Research Centerhttps://cyberbullying.org/facts.
Hinduja, S., & Patchin, J. W. (2019b). Connecting adolescent suicide to the severity of
bullying and cyberbullying. Journal of School Violence, 18(3), 333–346. https://doi.
org/10.1080/15388220.2018.1492417.
Hinshaw, S. P. (2013). Developmental psychopathology as a scientific discipline:
Rationale, principles, and advances. In T. P. Beauchaine, & S. P. Hinshaw (Eds.). Child
and adolescent psychopathology (pp. 3–27). (2nd ed.). New York: Wiley.
Kann, L., McManus, T., Harris, W. A., et al. (2018). Youth Risk Behavior
Surveillance—United States, 2017. Surveillance summary 2017:67 (no. 8). [inclusive
pp. 17-18]. Retrieved on June 14, 2019 from https://www.cdc.gov/healthyyouth/
8
Aggression and Violent Behavior 50 (2020) 101343
data/yrbs/pdf/2017/ss6708.pdf.
Kessel Schneider, S., Smith, E., & O’Donnell, L. (2013, November). Social media and cy-
berbullying: Implementation of school-based prevention efforts and implications for social
media approaches. Education Development Center. Retrieved on June 14, 2019 from
http://www.promoteprevent.org/social-media-and-cyberbullying-implementation-
school-based-prevention-efforts-and-implications.
Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Lattanner, M. R. (2014). Bullying in
the digital age: A critical review and meta-analysis of cyberbullying research among
youth. Psychological Bulletin, 140(4), 1073–1137. https://doi.org/10.1037/
a0035618.
Kowalski, R. M., Limber, S. P., & McCord, A. (2019). A developmental approach to cy-
berbullying: Prevalence and protective factors. Aggression and Violent Behavior, 45,
20–32. https://doi.org/10.1016/j.avb.2018.02.009.
Lenhart, A., Madden, M., Smith, A., Purcell, K., Zickuhr, K., & Rainie, L. (2011). Teens,
kindness and cruelty on social network sites. Washington, D.C: Pew Research Center.
Milosevic, T. (2016). Social media companies’ cyberbullying policies. International Journal
of Communication, 10, 5164–5185.
National Crime Prevention Council. (n.d.). What parents can do about cyberbullying.
Retrieved on June 17, 2019 from https://www.ncpc.org/resources/cyberbullying/
what-parents-can-do-about-cyberbullying/.
Olweus, D. (1999). Sweden. In P. K. Smith, Y. Morita, J. Junger-Tas, D. Olweus, R.
Catalano, & P. Slee (Eds.). The nature of school bullying: A cross-national perspective (pp.
7–27). London: Routledge.
Olweus, D., & Limber, S. (2018). Some problems with cyberbullying research. Current
Opinion in Psychology, 19, 139–143. https://doi.org/10.1016/j.copsyc.2017.04.012.
Patchin, J. W., & Hinduja, S. (2012). Cyberbullying: An update and synthesis of the re-
search. In J. W. Patchin, & S. Hinduja (Eds.). Cyberbullying prevention and response:
Expert perspectives (pp. 13–35). New York, NY: Routledge.
Perren, S., Corcoran, L., Cowie, H., Dehue, F., Garcia, D., Mc Guckin, C., et al. (2012).
Coping with cyberbullying: A systematic literature review. Final report of the COST IS
0801 working group 5 (published online).
Peter, I.-K., & Petermann, F. (2018). Cyberbullying: A concept analysis of defining at-
tributes and additional influencing factors. Computers in Human Behavior, 86,
350–366. https://doi.org/10.1016/j.chb.2018.05.013.
Pharo, H., Sim, C., Graham, M., Gross, J., & Hayne, H. (2011). Risky business: Executive
function, personality, and reckless behavior during adolescence and emerging
adulthood. Behavioral Neuroscience, 125(6), 970–978. https://doi.org/10.1037/
a0025768.
Rosa, H., Pereira, N., Ribeiro, R., Ferreira, P. C., Carvalho, J. P., Oliveira, S., ... Trancoso,
I. (2019). Automatic cyberbullying detection: A systematic review. Computers in
Human Behavior, 93, 333–345. https://doi.org/10.1016/j.chb.2018.12.021.
Rosenthal, S. R., Buka, S. L., Marshall, B. D. L., Carey, K. B., & Clark, M. A. (2016).
Negative experiences on Facebook and depressive symptoms among young adults.
Journal of Adolescent Health, 59(5), 510–516. https://doi.org/10.1016/j.jadohealth.
2016.06.023.
Savage, M. W., & Tokunaga, R. S. (2017). Moving toward a theory: Testing an integrated
model of cyberbullying perpetration, aggression, social skills, and internet self-effi-
cacy. Computers in Human Behavior, 71, 353–361. https://doi.org/10.1016/j.chb.
2017.02.016.
Selkie, E. M., Fales, J. L., & Moreno, M. A. (2016). Cyberbullying prevalence among US
middle and high school–aged adolescents: A systematic review and quality assess-
ment. Journal of Adolescent Health, 58(2), 125–133. https://doi.org/10.1016/j.
jadohealth.2015.09.026.
Slotter, E. B., & Finkel, E. J. (2011). I3 theory: Instigating, impelling, and inhibiting
factors in aggression. In P. R. Shaver, & M. Mikulincer (Eds.). Herzilya series on per-
sonality and social psychology. Human aggression and violence: Causes, manifestations,
and consequences (pp. 35–52). Washington, DC, US: American Psychological
Association. https://doi.org/10.1037/12346-002.
Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008).
Cyberbullying: Its nature and impact in secondary school pupils. Journal of Child
Psychology and Psychiatry, 49, 376–385. https://doi.org/10.1111/j.1469-7610.
(2007).01846.x.
Smith-Spark, L. (2013, August 9). Hanna Smith suicide fuels calls for action on Ask.fm
cyberbullying. Retrieved on June 14, 2019 from http://www.cnn.com/2013/08/07/
world/europe/uk-social-media-bullying/.
Suler, J. (2004). The online disinhibition effect. Cyberpsychology & Behavior, 7(3),
321–326. https://doi.org/10.1089/1094931041291295.
Swearer, S. M., & Hymel, S. (2015). Understanding the psychology of bullying: Moving
toward a social-ecological diathesis-stress model. American Psychologist, 70(4),
344–353. https://doi.org/10.1037/a0038929.
Tanrikulu, I. (2018). Cyberbullying prevention and intervention programs in schools: A
systematic review. School Psychology International, 39(1), 74–91. https://doi.org/10.
1177/0143034317745721.
United States Department of Health and Human Services (2018, January). Laws, policies
and regulations. Retrieved on June 14, 2019 from https://www.stopbullying.gov/
laws/index.html.
Vaillancourt, T., Faris, R., & Mishna, F. (2017). Cyberbullying in children and youth:
Implications for health and clinical practice. Canadian Journal of Psychiatry, 62(6),
368–373. https://doi.org/10.1177/0706743716684791.
Valencia, N. (2014, March 10). Yik Yak chat app stirring up trouble in high schools.
Retrieved June 14, 2019 from http://www.cnn.com/2014/03/07/tech/yik-yak-app-
high-school-problems/.
Van Cleemput, K., DeSmet, A., Vandebosch, H., Bastiaensens, S., Poels, K., & De
Bourdeaudhuij, I. (2014). A systematic review and meta-analysis of the efficacy of
cyberbullying prevention programs. Paper presented at ‘Etmaal van de
Communicatiewetenschap’, Wageningen, NL Feb 3rd-4th.
N.S. Ansary
Vandebosch, H., & Van Cleemput, K. (2009). Cyberbullying among youngsters: Profiles of
bullies and victims. New Media Society, 11, 1349–1371. https://doi.org/10.1177/
1461444809341263.
Waasdorp, T. E., & Bradshaw, C. P. (2015). The overlap between cyberbullying and tra-
ditional bullying. Journal of Adolescent Health, 56(5), 483–488. https://doi.org/10.
1016/j.jadohealth.2014.12.002.
Wallace, K. (2015, January 9). Parents, beware of bullying on sites you've never seen.
Retrieved June 14, 2019 from http://www.cnn.com/2013/10/10/living/parents-
new-apps-bullying/.
Whittaker, E., & Kowalski, R. (2015). Cyberbullying via social media. Journal of School
Aggression and Violent Behavior 50 (2020) 101343
Violence, 14(1), 11–29. https://doi.org/10.1080/15388220.2014.949377.
Wong, R. Y. M., Cheung, C. L. K., & Xiao, B. (2018). Does gender matter in cyberbullying
perpetration? An empirical investigation. Computers in Human Behavior, 79, 247–257.
https://doi.org/10.1016/j.chb.2017.10.022.
World Health Organization [WHO] (2013). Definition of key terms. Retrieved from
https://www.who.int/hiv/pub/guidelines/arv2013/intro/keyterms/en/.
Zych, I., Farrington, D., & Ttofi, M. M. (2019). Bullying and cyberbullying: Protective
factors and effective interventions. Aggression and Violent Behavior, 45, 4–19. https://
doi.org/10.1016/j.avb.2018.06.008.